Today's optical fiber networks carry many channels over long distances. When a signal is transmitted via an optical fiber, the signal's power level degrades the longer the transmission distance becomes, as the incident photons interact with glass fiber molecules and/or with photons from adjacent wavelengths. Power level degradation is a significant challenge in maintaining optical fiber networks; manifesting as phenomena such as chromatic dispersion, self-phase modulation, four wave mixing, cross-phase modulation, stimulated brilloin scattering, stimulated raman scattering, and polarization mode distortion.
To compensate for power level degradation, nodes having light amplifiers are provided at predetermined intervals to amplify the signal before transmitting it on to the next fiber span. Existing methods of power level adjustment, or “equalization” as it is referred to in Dense Wave Division Multiplexing (DWDM) systems, involve a process whereby channels are individually adjusted using non-real time algorithms to optimize overall system performance.
Measured values for variables such as current channel input power, amplifier noise level, current transmission setting, optical signal to noise ratio (OSNR), and the like are input into an equalization algorithm. The output provides a resultant recommended equalization factor, or transmission power adjustment value.
The recommended transmission power adjustment value is typically capped by what is known in the industry as a “clamp”. Existing clamps are fixed values that limit the maximum power adjustment on any given iteration of the algorithm. A large clamp will result in a fast, sub-optimal convergence, and a small clamp will result in a slow, optimal convergence, so that using a fast convergence of the equalization algorithm is typically at the expense of optimality, while optimal convergence of the algorithm is typically at the expense of speed.
The problem is that existing clamping methods are a compromise between optimality and speed. What is needed is a way to divorce these two conflicting objectives so that equalization can be performed with both optimality and speed.
For the foregoing reasons, there is a need for an improved method of equalization.